Educators and researchers have examined the effectiveness of grade retention for decades. From the early 1980s to the present, opinions regarding the merit of grade retention have varied from being positive for some students and in some circumstances, to being overwhelmingly negative and of little value in promoting academic achievement in others. Several conclusions regarding the helpfulness and harmfulness of retention can be drawn from the wealth of research that has been conducted. Overall, one point is clear—retention is not effective in producing significant gains in student achievement or in having lasting benefits for struggling students (Alexander et al., 1994; Shepard & Smith, 1989; Thompson, 1999).

Prevalence and Cost of Grade Retention

The prevalence of grade retention, much like social promotion, is unknown, because school districts rarely keep records of how many children are retained each year. Estimates can be derived, however, from census data. These estimates show that the number of grade-retained children ranges from six to nine percent annually (Association of California School Administrators, 1999; Center for Policy and Research in Education [CPRE], 1990). For students in urban school districts, the retention rate has been estimated to be approximately 50 percent (AFT, 1997). The cost of retaining or reeducating U.S. students for at least one year is staggering. For example, in 1996 to 1997 there were about 46 million children enrolled in public schools in the United States with an average cost per pupil expenditure of$5,923 (NCES, 1999b). Using these figures, this means that at least 3.2 million children (7 percent) were retained in grade at a cost of almost $19 billion.

Retention and Academic Achievement

Retention can help some students and in certain circumstances, but there are serious risks associated with it (Thompson, 1999). The majority of studies show that retention is not effective in promoting positive academic achievement, especially in the long run. Holmes (1989) conducted a meta-analysis of 63 empirical studies that examined the effectiveness of retention. Fifty-four of the 63 studies showed that at-risk children who were promoted achieved at the same or higher levels than comparable peers who were retained and spent two years rather than one in a grade. Other studies have also confirmed these findings and found that when retained and promoted students of like ability were compared, the promoted students outperformed the retained students the next year (Norton, 1990; Walters & Borgers, 1995).

Children are often retained in kindergarten or first grade in the belief that if a student must be retained, it is best to do so very early in a child's school career. Several well-designed studies show, however, that retaining children in kindergarten and first grade ultimately can be ineffective and harmful. A large study in the Chicago public schools showed that retained children, especially in first grade, did not improve over time (Reynolds, Temple, & McCoy, 1997). Alexander and colleagues (1994) followed 775 students in Baltimore city schools for eight years. They found that children retained in first grade improved their achievement test scores the year they were retained. However, these same retainees achieved in second grade and every grade thereafter at the same relative level as their first year in first grade. Researchers speculate first grade retention may be especially difficult for young children because of the difficulty they experience transitioning into a formal school environment. Being removed from peers with whom they have formed a relationship appears to hinder their development rather than foster positive growth (Entwisle & Alexander, 1993).

There is a considerable amount of research that also shows that kindergarten retention and use of transition grades (such as two-year kindergarten programs) have no lasting academic advantage over other children who were never retained but were also equally unready because of immaturity or low achievement. Children who spent the extra year in kindergarten were just as likely as their promoted counterparts to be at the bottom of the third grade class (Gredler, 1984; Holmes, 1989; Meisels, 1992; Nason, 1991; Rose, Medway, Cantrell, & Marus, 1983; Shepard & Smith, 1986, 1989).

Retention and Social and Health Implications

Children who are retained tend to feel more poorly about their capabilities, score lower on measures of personal and psychological adjustment, and display more discipline problems. Clinical interviews with students show that they felt angry or sad about the retention and feared the reaction of family and friends. Some were teased by neighbors and reported having a difficult time adjusting to school (Byrnes, 1989; Holmes, 1989; Norton, 1990; Shepard & Smith, 1989).

Being retained in a grade has also been strongly correlated with dropping out of school. Children who are retained one year are five times more likely to drop out of school than those who have never been retained. Children who are retained two or more years have almost a 100 percent probability of becoming dropouts compared with similar low performers who are promoted (CPRE, 1990).

School failure has also been linked to participation in health-risk behaviors (cigarette use, alcohol use, and weapons-related violence) for adolescents, according to data from the National Longitudinal Study of Adolescent Health (Blum, Beuhring, Shew, Bearing, Sieving, & Resnick, 2000). In-home interviews with a nationally representative sample of about 10,000 students ages 12 to 17 and their families showed that school failure is more likely to predict participation in violent activities, use of alcohol, and involvement in sexual activity than is poverty, race, or family structure. The researchers concluded that school failure should be viewed as a public health problem (Blum et al., 2000).

Conclusions about Retention

Even though a few studies have found that retention can have a positive short-term benefit (Alexander et al., 1994; Holmes, 1989), the vast majority have shown either no long-term advantage, harm, or a consistent "washout" effect (Gredler, 1984; Holmes, 1989; Mantizicopoulos & Morrison, 1992; Meisels, 1992; Nason, 1991; Reynolds et al., 1997; Rose et al., 1983; Shepard & Smith, 1986, 1989). This indicates that the potential benefit does not warrant the risk. Even researchers who otherwise support retention do not view it as a tool for helping children succeed in school (Alexander et al., 1994).